When a computer with a browser is connected as a client to a server by way of the internet, the server ordinarily provides pages or screens on request, together with the logic required to generate and process the screen image. A command is given at the browser-enabled computing device or user terminal, either as a keystroke or as a mouse click. Many, but not all, of these commands or requests are sent to the server. Most commonly, the server and browser-enabled computing device are remote from each other, and the communication is by way of the Internet. The server generates a response, which may be a new page, fill in a table on the current page, or the like. The new page or information is displayed by the browser.
Repetitive commands may occur, either because of some mechanical difficulty such as an object lying on a keypad or because of an impatient user. These repetitive commands are second or later commands that request actions of the server which have not yet been fulfilled the first time by the server.
The result of repetitive requests is the possibility of at least slowing the operation of the server as it attempts to handle the requests. This in turn leads to more commands occurring during the delay. It also results in the possibility of “hanging” or stalling (crashing) the server operation. Another problem with repetitive requests is that the user may lose track of the sequence of commands that has been entered, and may therefore not be aware of the end result of the sequence of commands already entered.
A disadvantageous effect of server response to repetitive requests is that, if many requests of the same sort are commanded in a short period of time, the server will process those requests in sequence, so that the requested actions may occur repeatedly after the commands are no longer being given.
Improved control is desired over commands sent from a browser to a server.